Oracle Linux 7 Desktop Refreshes Periodically

Question

Hi:
I have a CentOs 6.6 system (lots of CPUS, lots of RAM) running VirtualBox.  Some time ago I had installed a VM from an Oracle Developer Day which had Oracle Linux 6 on it along with Oracle 12c DB, SQL Developer, etc.  That's been running fine.  I noticed a new VM that was Oracle Linux 7 and Oracle DB 12c, SQL Developer etc. so I got that image and imported it, creating a new Oracle Linux 7 VM.
It comes up fine but every few minutes it does a refresh of some kind involving the entire desktop.  Everything freezes for a few seconds, then the red image of the background turns to black (but I still see all the icons), then entire screen goes gray (no icons), and usually things come back.  Usually.  But sometimes I am greeted with "Oh no!  Something has gone wrong...and the system can't recover".  The machine isn't dead but I have to log in again and my sessions are gone.
I've only had the machine up for a half hour or so, so I don't know if it is certain activity that is driving the refresh or what. This is a major pain.  Is this a known issue?  How do I get rid of it?
Thanks
P.S. The VM image in question can be found here.

Answer 1

Hi Dude!  Thanks for replying.
Resources are NOT an issue. The host has 18 Xeon E5-2650 CPUs and 128G or so of RAM   The guest is getting 6 cores and 12G of RAM.
Here are some details as to the VM properties:
Oracle DB Dev VM:
Base Memory: 12154MB
Processors: 6
Enable I/O APIC
Enable Hardware Clock in UTC Time
Enable PAE/NX
Enable VT-x/AMD-V
Enable Nested Paging
No Video acceleration
Boot up shows: Fast TSC Callibration Fails
Could giving this VM 6 cores be a problem?  I don't know how Oracle DB handles that if the VM was created using 1 core and I've upped it to 6.  Does it not use them all?  Does it get confused?
I looked at /var/log/message and there are two things.  It's full of problems regarding time.
Jan  1 14:46:57 localhost abrt-server: Missing file: time
Jan  1 14:46:57 localhost abrt-server: Unlocked '/var/tmp/abrt/ccpp-2014-08-15-14:26:33-1953/.lock' (no or corrupted 'time' file)
I think I mentioned in the first post that on startup it always has an error about the time.  It shows Fast TSC calibration failed.
But I think that's a general RedHat7/CentOs7 guest thing from what I have read and it isn't critical and certainly shouldn't cause this problem.
The other thing I noticed in the message log around the time of the last crash is a line as follows:
Jan  1 14:26:32 localhost dbus[805]: [system] Successfully activated service 'org.freedesktop.PackageKit'
Jan  1 14:28:26 localhost kernel: gnome-shell[3261]: segfault at 3ccea774 ip 00007f4392fa8690 sp 00007f43b8deb770 err
or 4
Jan  1 14:28:28 localhost abrt-hook-ccpp: Saved core dump of pid 3240 (/usr/bin/gnome-shell) to /var/tmp/abrt/ccpp-20
15-01-01-14:28:26-3240 (407973888 bytes)
Jan  1 14:28:29 localhost abrt-server: Missing file: time
Jan  1 14:28:29 localhost abrt-server: Unlocked '/var/tmp/abrt/ccpp-2014-07-31-22:08:30-1470/.lock' (no or corrupted
'time' file)
So something (gnome related?) died but I don't see a hint as to why.  I didn't see anything in the Xorg file that stood out.
FWIW, it's happening a bit less since I did a yum update on the machine.